The backbone of an antibody, known as the Fc region, is responsible for pharmacokinetic properties that may be desirable in the case of many therapeutic biologics (Jeffries, B. Biotechnol Prog. 2005; 21: 11-16). The size of the Fc region makes it resistant to renal filtration and binding to the Fc Neonatal Receptor (FcRn) allows it to escape endosomal degradation by a recycling mechanism. In addition to the Fc region that is present in monoclonal antibody therapeutic products, there are Fc fusion products being investigated and developed (Hakim et al. Mabs. 2009; 1:281-287). Fc fusions are the fusion of an Fc region to another protein, peptide, or Active Pharmaceutical Ingredient (API). The Fc fusion then has both the properties of the Fc region and the therapeutic properties of the API.
There are many cell lines that are capable of being used to manufacture therapeutic biologics (Jung et al. Curr Opin Biotechnol. 2011; 22:1-10). The mammalian Chinese Hamster Ovary (CHO), insect Sf9, yeast S. cereviae, and bacterial E. coli are some of the most common cell lines that are discussed for recombinant protein production. So far, yeast, CHO and E. coli have been used for manufacture of Fc containing therapeutic biologics, including a large number of monoclonal antibodies. Expression in E. coli offers three potential and significant advantages over expression in other cell lines: the cell line development time is much shorter; the bioreactor runs are up to 7-fold shorter, resulting in a lower capital investment; and there is no need to control aberrant glycosylation that can occur in yeast and mammalian cell cultures.
Expression of larger proteins, like Fc fusions, in E. coli can be a unique challenge. E. coli lack the chaperone proteins and other refolding machinery found in a eukaryotic expression system. The cytoplasm of E. coli is also a reducing environment, which is not favorable for the formation of disulfide bonds. The Fc region of human IgG1 antibodies contains six disulfide bonds. Two disulfide bonds the hinge region join two peptide chains to form the homodimeric molecule and there are two more disulfide bonds within each of the peptide chains. E. coli also have a mechanism to prevent unfolded proteins from interfering with normal cell processes. Unfolded protein is shunted and isolated in insoluble aggregates, called Inclusion Bodies (IB), which can then be isolated in the insoluble fraction following cell lysis. Alternatively, when the rate of recombinant protein production is slowed to allow the protein to fold, a leader sequence may be added to direct soluble protein that is expressed to the periplasmic space. The periplasm is an oxidative environment favorable for the formation of disulfide bonds. However, the reported expression levels of recombinant protein in the periplasm remain low (Liu et al. Protein Expression Purif. 2008; 62:15-20).
In contrast, E. coli expression levels in IBs have been reported to be high. Expressing protein in IBs also has the advantages of resistance to protein degradation, and ease of isolation from the cells (Grune et al. Int J Biochem Cell Biol. 2004; 36:2519-2530). Since an IB is an insoluble aggregate, there may be a challenge in restoring the protein of interest to its biologically active conformation (Jungbauer et al. J Biotechnol. 2006; 587-596). Typically, a process is required to break apart and solubilize the IB. Then the protein must be renatured, or refolded, into the biologically active conformation while minimizing losses due to aggregation and precipitation. Current refolding processes may be specific to a given protein, requiring thorough optimization for each case. Many refolding processes require very low protein concentrations and consequently large volumes for the operation. This is difficult because it requires a larger amount of potentially expensive reagents. There is also a challenge in a manufacturing setting, where there is a physical limit to the container size that may be used to refold proteins. Finally, in the case of Fc fusions, the refolding process must correctly form the six disulfide bonds that exist in the native form of the protein.